CCD or CMOS image sensors for consumer digital still ... · CCD or CMOS image sensors for consumer...

1

2001 Albert J.P. Theuwissen, Philips Semiconductors Image Sensors

CCD or CMOS image sensorsCCD or CMOS image sensors

for consumer digital still for consumer digital still

photography ?photography ?

Prof. dr. ir. Albert J.P. THEUWISSENPhilips Semiconductors Image Sensors, Eindhoven (NL)

Technical University, Delft (NL)

[email protected]


Outline

• Introduction

• Principle of CCD and CMOS imagers

• Imager requirements

• Overview CCD vs. CMOS : resolution, signal-to-noise ratio, angular response, dark

current, dynamic range, linearity, pixel

uniformity, architecture

• Summary and Conclusions

- introduction

- principle ofimagers

- imager requirements

- overview CCD vs. CMOS

- summary andconclusions

2


Introduction

• CMOS is challenging CCD

• Digital still is a continuously growing imaging

market

- introduction






Introduction

• CMOS is challenging CCD

• Digital still is a continuously growing imaging

market

• Today : almost exclusively CCD in DSC

• Tomorrow : CCD or CMOS ?

- introduction





3


Outline

• Introduction







- introduction






CCD principle (1)

0 V

p-Si

10 V 0 V 0 V

0 V

p-Si

10 V 10 V 0 V

0 V

p-Si

0 V 10 V 0 V

ΦΦΦΦ

ΦΦΦΦ

ΦΦΦΦ

- introduction





- CCD

4


CCD principle (2)

out horizontal CCD output register

photosensitive CCD array

- introduction





- CCD


CMOS principle (1)

vertical scan circuit

photodiode array + MOS switches

horizontal scan circuit

A/D

- introduction





- CCD

- CMOS

5


CMOS principle (2)

p-Sin+

Photogate APS

RS

Col bus

RST

TXPG

p-Sin+

Photodiode APS

RS

Col bus

RST

- introduction





- CCD

- CMOS


Outline

• Introduction







- introduction





- CCD

- CMOS

6


Image Sensor Aspects (1)

IMAGER

PARAMETER

resolution

signal-to-noise ratio

angular response

dark current

CAMERA

SPECIFICATION

sharpness

ISO speed

min. F-stop

max. exp. time

- introduction





- CCD

- CMOS


Image Sensor Aspects (2)

IMAGER

PARAMETER

dynamic range

linearity

pixel uniformity

architecture

CAMERA

SPECIFICATION

latitude

colour fidelity

granularity

features

- introduction





- CCD

- CMOS

7


Outline

• Introduction







- introduction





- CCD

- CMOS


Resolution Requirements

0

200

400

600

0 2 4 6 8 10

Image Size [Mpixels]

Print Resolution [pix/inch]

3.5"x5"

4"x6"

5"x7"

8"x10"quality level

- introduction





- CCD

- CMOS

- resolution

8



0

200

400

600

0 2 4 6 8 10



3.5"x5"

4"x6"

5"x7"

8"x10"

disc 110 APS 35 mm

quality level

- introduction





- CCD

- CMOS

- resolution



0

200

400

600

0 2 4 6 8 10



3.5"x5"

4"x6"

5"x7"

8"x10"

disc 110 APS 35 mm

cons. CMOS

cons. CCDprof. CCD

quality level

prof. CMOS

- introduction





- CCD

- CMOS

- resolution

9


Trend in Resolution

0

2

4

6

8

10

0 2 4 6 8 10

Pixel Size [um]

Resolution [Mpixels] 2/3" (8.8mmx6.6mm)

1/2" (6.4mmx4.8mm)

1/3" (4.4mmx3.3mm)

decrease chip size

- introduction





- CCD

- CMOS

- resolution


Trend in Resolution

0

2

4

6

8

10

0 2 4 6 8 10

Pixel Size [um]

Resolution [Mpixels] 2/3" (8.8mmx6.6mm)

1/2" (6.4mmx4.8mm)

1/3" (4.4mmx3.3mm)

decrease chip size

lens limitation

CCDCMOS

- introduction





- CCD

- CMOS

- resolution

10


Signal-to-Noise ratio (1)

0.001 0.01 0.1 1 10 10010

100

1k

100kNumber of Electrons

Light Intensity (µµµµW/cm )

10k

photon shot

read noise (20)

total (40)darkshot (35)

signal

saturation (100k)

dynamic range (68 dB)

2

- introduction





- CCD

- CMOS

- resolution

- S/N ratio



p-Sin+

Photogate APS

RS

Col bus

RST

TXPG

p-Sin+

Photodiode APS

RS

Col bus

RST

NOT FREE of reset noiseFREE of reset noise

LOW light sensitivity HIGH light sensitivity

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

11



QEAISO40

⋅∝

ISOx = ISO-speed @ S/N=x

Hx = exposure to get S/N=x

A = pixel area

QE = quantum efficiency

nr = read noise

x

xH

10ISO =

r

10n

QEAISO

⋅∝

- introduction





- CCD

- CMOS

- resolution

- S/N ratio



0

20

40

60

80

100

0 1 2 3 4 5 6 7 8 9 10

Pixel Size [um]

QE in green [%]

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

12


0

20

40

60

80

100

0 1 2 3 4 5 6 7 8 9 10

Pixel Size [um]

QE in green [%]

CCD


CCD

CMOS

- introduction





- CCD

- CMOS

- resolution

- S/N ratio


Improvement QE (1)

micro-lenspixels

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

13


Improvement QE (2)

lens

iris

micro-lenspixels

- introduction





- CCD

- CMOS

- resolution

- S/N ratio


Angular Response (1)

0

5

10

15

20

25

1.2 1.4 1.8 2 2.8 3.5 4 5.6 8 16

F-number

Incident Angle [deg.]

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

14



0

0.2

0.4

0.6

0.8

1

0 5 10 15 20 25 30


Rel. Imager Response

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.



0

5

10

15

20

25

1.2 1.4 1.8 2 2.8 3.5 4 5.6 8 16

F-number


micro-lens limit

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

15


Dark Current (1)

• prof. CCD

• cons. CCD

• standard CMOS

Dark current doubles every 6…8oC.

Example : @ 60oC : 32 times higher !

@ -100oC : 32,000 times lower !

1...3 pA/cm2 @ RT

5...10 pA/cm2 @ RT

200...500 pA/cm2 @ RT

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current


Dark Current (2)

• Compensation for dark current is possible !

• Compensation for dark-current non-

uniformities is possible !

• Compensation for dark-current shot-noise is

NOT possible !

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

16


Dynamic Range (1)

2

dark

2

r

darksat

nn

NNDR

+

−=

DR = dynamic range

Nsat = saturation signal [e-]

Ndark = dark signal [e-]

nr = read noise [e-]

ndark= dark shot noise [e-]

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range


Dynamic Range (2)

0

0.5

1

1.5

2

2.5

3

2 3 4 5 6 7 8 9 10

Pixel Size [um]

Charge Handling [ke/um2]

Nsat = 30 ke-

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

17


Dynamic Range (2)

0

0.5

1

1.5

2

2.5

3

2 3 4 5 6 7 8 9 10

Pixel Size [um]

Charge Handling [ke/um2]

CMOS

CCD

Nsat = 30 ke-

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range


Linearity (1)

RAWdata

inter-polation

whitebalance

colourmatrixing

gammacurve

RGBdata

R = R x 1.40G = G x 1.00B = B x 1.46

RGB

1.424-0.0460.090

-0.5001.431-0.390

0.076-0.4911.302

RGB

= xγγγγ = 1.8

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

18


Linearity (2)

• Due to sampling in colour space :

Interpolations,

• Filters do not match perfectly : Colour

corrections.

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity


Linearity (2)

• Due to sampling in colour space :

Interpolations,

• Filters do not match perfectly : Colour

corrections.

• Linearity CCD : 99 % (for 70 % of Nsat),

• Linearity CMOS : 97 % (for 85 % of Nsat).

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

19


Pixel Random Non-

Uniformity

• PRNU CCD :

• PRNU CMOS :

< 0.7 … 1.0 %,

< 2.0 … 5.0 %,

(column + pixel FPN)

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.


Pixel Random Non-

Uniformity

• PRNU CCD :

• PRNU CMOS :

• Can be corrected by means of LUT,

• To be non-visible : PRNU and FPN < photon shot

noise (0.5 % for 40 ke-).

< 0.7 … 1.0 %,

< 2.0 … 5.0 %,

(column + pixel FPN)

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

20


Architecture

• CCD : parallel integration/reset

• CMOS : rolling integration/reset

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture


Architecture

• CCD : parallel integration/reset

• CMOS : rolling integration/reset

• Can be solved by 1 T and 1 C extra in every

pixel extra …

• Costs sensitivity, charge capacity, noise, ...

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture

21


Outline

• Introduction







- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture


Summary (1)

• Resolution : pixel size of CCD smallest

• Noise : CMOS pixels suffer from reset noise

• Quantum efficiency : CMOS and CCD can be

similar

• Angular response : limits set by micro-lenses

• Dark current : CCD outstanding

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture

- summary

22


Summary (2)

• Saturation level can be similar

• Dynamic range of CCD is higher

• Linearity of CCD is better

• Pixel uniformity of CCD is better

• Device architecture of CCD gives more

flexibility

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture

- summary


Conclusions (1)

CCD or CMOS image sensor

for consumer digital still

photography ?

CCD ? YES !

CMOS ? YES, provided that noise and

dark current problems can be solved !!!

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture

- summary

- conclusions

23


Conclusions (2)

• Main issue : S/N performance

• Then benefit from :

– low power of CMOS,

– low driving voltages of CMOS,

– on-chip functionality,

– selective read-out mechanism,

– cost advantage.

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture

- summary

- conclusions


Important Remark

This presentation was about digital

still photography.

For video applications the situation

changes completely !!!

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture

- summary

- conclusions

- remark

24


References

• R. Baer : IEEE workshop on CCD & AIS, Karuizawa,

1999,

• J. Bosiers et.al. : IEDM, San Francisco, 1998,

• M. Kriss : ICPS, Antwerp, 1998,

• A. Theuwissen : “Solid-state imaging with Charge-

Coupled Devices”, 1995

• E. Fossum : “Camera-on-a-chip”, IEEE-ED, Oct.,

1997.

- introduction





- CCD

- CMOS

- resolution

- S/N ratio

- angular resp.

- dark current

- dyn. range

- linearity

- p.r.n.u.

- architecture

- summary

- conclusions

- remark

- references

CCD or CMOS image sensors for consumer digital still ... · CCD or CMOS image sensors for consumer...

Documents

Transcript of CCD or CMOS image sensors for consumer digital still ... · CCD or CMOS image sensors for consumer...